Vehicle driving control device

Abstract

To reduce the risk of traffic flow disturbance at an intersection or contact / collision with other traffic participants during operation of a risk mitigation function.SOLUTION: In a vehicle 1 having an automatic operation device with remote monitoring / operation, which performs a risk mitigation function (RMF) for stopping the vehicle 1 at a target stop position when it is difficult to continue the remote monitoring / operation, on the basis of location information and map information on the vehicle 1, in preparation for operating the RMF, target stop position candidates are searched, and when the RMF is operated near an intersection, for each of the plurality of target stop position candidates including a target stop position candidate located on a route different from an automatic running target route PA, a risk factor RF which indicates a level of risk the vehicle 1 has from a current position to the target stop position candidate is calculated, and a target stop position candidate with the lowest risk factor RF is selected as a target stop position from among the plurality of target stop position candidates.SELECTED DRAWING: Figure 7

Description

[Technical Field]

[0001] The present invention relates to a vehicle driving control device, and more particularly to a risk reduction function for a vehicle during automatic driving through remote monitoring and operation. [Background technology]

[0002] Development is underway to develop technologies that will enable unmanned vehicles to be driven under specific conditions using remotely monitored and operated automated driving devices. Remotely operated automated driving systems are configured to monitor the driving of a vehicle operated by an automated driving device equipped with, for example, an Accelerated Control System (ACCS) or a continuous automatic steering system from a remote control base station and operate the vehicle as necessary. In such remotely operated automated driving systems, if for some reason it becomes difficult to continue remote monitoring and operation while the vehicle is being driven by the automated driving device, it is necessary to respond using the risk mitigation function (RMF) of the automated driving device installed in the vehicle.

[0003] For example, Patent Document 1 discloses that when an abnormality occurs in an autonomously driven vehicle, an evacuation site is searched for, a taxiway to the evacuation site is calculated, and the vehicle is controlled to travel along the taxiway by automatic steering. As the evacuation site, a space where the vehicle can be parked, such as a vacant lot or a parking lot of a commercial facility, is set. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] Japanese Patent Application Publication No. 2019-152963 Summary of the Invention [Problem to be solved by the invention]

[0005] However, if the RMF is activated near an intersection, there is a risk that the vehicle may cross paths with other traffic participants at the intersection, causing traffic flow disruptions and contact or collision with other traffic participants while guiding the vehicle to an evacuation location.

[0006] The present invention has been made in consideration of the above-described circumstances, and its purpose is to reduce the risk of traffic flow disruption at intersections and contact or collision with other traffic participants when the risk mitigation function is activated. [Means for solving the problem]

[0007] Book According to another aspect of the invention, there is provided a driving control device for a vehicle equipped with an automatic driving device for performing automatic driving by remote monitoring and operation, the driving control device having a risk mitigation function (RMF) that performs risk mitigation control to stop the vehicle at a target stopping position when it becomes difficult to continue remote monitoring and operation, the automatic driving device constantly searches for a plurality of target stopping position candidates based on position information and map information of the vehicle during the automatic driving in preparation for activation of the RMF, and when the RMF activates near an intersection, searches for a plurality of target stopping position candidates including target stopping position candidates located on a route different from the target route of the automatic driving, For each of the candidate locations, a risk factor representing the level of risk involved in the vehicle reaching the candidate target stop location from its current location is calculated, and the candidate target stop location with the lowest risk factor is selected as the target stop location from among the plurality of candidate target stop locations. The risk factor is calculated as the sum of points related to the number of intersections between the vehicle and other traffic participants at the intersection and in the vicinity of the intersection, points related to the predicted travel time from the current location of the vehicle to the candidate target stop location, and points related to the predicted travel distance from the current location of the vehicle to the candidate target stop location. [Effects of the Invention]

[0008] The vehicle cruise control device according to the present invention can reduce the risk of disruption of traffic flow at an intersection and contact or collision with other traffic participants when the risk reduction function is activated near an intersection. [Brief explanation of the drawings]

[0009] [Figure 1] FIG. 1 is a schematic diagram showing a vehicle cruise control system according to an embodiment of the present invention. [Figure 2] FIG. 2 is a schematic plan view showing a group of external sensors of a vehicle. [Figure 3] FIG. 3 is a block diagram showing a vehicle driving control system. [Figure 4] 4(a) and 4(b) are diagrams illustrating the relationship between the stopping position of a vehicle and other traffic participants around an intersection when turning right. [Figure 5] 5(a) and 5(b) are diagrams illustrating the relationship between the stopping position of a vehicle and other traffic participants around an intersection when turning left. [Figure 6] 6(a) and 6(b) are diagrams illustrating the relationship between the stopping position of a vehicle and other traffic participants around an intersection when traveling straight. [Figure 7] 7(a) and (b) are diagrams showing a plurality of target stop position candidates set for RMF. [Figure 8] 8(a) and 8(b) are diagrams showing a plurality of target stop position candidates set for RMF. [Figure 9] FIG. 9 is a diagram for explaining the selection of a target stopping position at a T-junction. [Figure 10] FIG. 10 is a diagram for explaining the selection of a target stopping position at a crossroads. [Figure 11] FIG. 11 is a flowchart illustrating the flow of the RMF operation. DETAILED DESCRIPTION OF THE INVENTION

[0010] An embodiment of the present invention will be described in detail below with reference to the drawings. A cruise control system for a vehicle 1 according to this embodiment is configured to perform all driving operations within an operational design domain (ODD) and to be capable of performing automated driving equivalent to SAE (Society of Automotive Engineers) Level 4. The cruise control system for the vehicle 1 is a remote-controlled automated driving system capable of automated driving through remote monitoring and operation, and the vehicle 1 is envisioned to be a service car such as a taxi or rental car that provides an unmanned automated driving transportation service.

[0011] 1 to 3, vehicle 1 equipped with a cruise control system according to this embodiment includes, in addition to typical automobile components such as an engine and a body, external sensors 21 for detecting the environment around the vehicle, internal sensors 22 for detecting vehicle information, a map information database 23, positioning means 24, a group of controllers / actuators for speed control and steering control, an ACC controller 15 for controlling the distance between vehicles, an automatic steering controller 16 for automatic steering control, and an automatic driving device 10 for integrating these components to perform route tracking control, in order to perform the recognition, judgment, and operation that have traditionally been performed by a driver on the vehicle side. Vehicle 1 further includes a communication device 25 for communicating remote control commands and vehicle information, vehicle position information, etc. with a remote control base station 25R.

[0012] The controller / actuator group for speed control and steering control includes an EPS (electric power steering) controller 31 for steering control, an engine controller 32 for acceleration / deceleration control, and an ESP / ABS controller 33. ESP (registered trademark; Electronic Stability Program) includes ABS (anti-lock braking system) to form a stability control system (vehicle behavior stabilization control system).

[0013] The external environment sensor 21 consists of multiple detection means for inputting the presence and relative distance of road dividing lines that define the current lane and adjacent lanes, other vehicles, obstacles, people, etc. around the current vehicle as external environment data such as image data or point cloud data into the automatic driving device 10.

[0014] For example, as shown in Fig. 2, the vehicle 1 is equipped with a millimeter-wave radar (211) and a camera (212) as forward detection means 211, 212, a LIDAR (laser image detection / ranging) as front-side detection means 213 and rear-side detection means 214, and a camera (back camera) as rear detection means 215. The external sensor 21 covers 360 degrees around the vehicle and is capable of detecting the positions and distances of other vehicles and obstacles within a predetermined range in each of the front, rear, left, and right directions of the vehicle, as well as the positions of lane markings in the vehicle's lane and adjacent lanes. Note that millimeter-wave radar (or LIDAR) can also be added as rear detection means.

[0015] The internal sensor 22 is made up of a plurality of detection means for measuring physical quantities that represent the motion state of the vehicle, such as a vehicle speed sensor, a yaw rate sensor, an acceleration sensor, etc. As shown in Fig. 3, the measured values ​​of the internal sensor 22 are input to the automatic driving device 10, the ACC controller 15, the automatic steering controller 16, and the EPS controller 31, and are processed together with the input from the external sensor 21.

[0016] The automatic operation device 10 includes an environmental state estimation unit 11, a risk mitigation function (RMF) unit 12, a route generation unit 13, and a vehicle control unit 14, and is composed of a computer for performing the functions described below, namely, a ROM that stores programs and data, a CPU that performs arithmetic processing, a RAM that reads the programs and data and stores dynamic data and arithmetic processing results, and an input / output interface.

[0017] The environmental state estimation unit 11 acquires the absolute position of the vehicle by matching the vehicle position information obtained by positioning means 24, such as a Global Navigation Satellite System (GNSS), with the map information in a map information database 23, and estimates the positions of lane markings in the vehicle's lane and adjacent lanes, and the positions and speeds of other vehicles, based on external data such as image data and point cloud data acquired by an external sensor 21. The environmental state estimation unit 11 also acquires the vehicle's motion state from internal data measured by an internal sensor 22.

[0018] The RMF unit 12 determines whether to activate a risk mitigation function (RMF) that safely stops the vehicle 1 within a target stopping area when road conditions or environmental conditions are outside the operation design domain (ODD) or when the system does not operate normally, based on information input from the environmental state estimation unit 11. In this embodiment, the RMF is configured to activate when it becomes difficult to continue remote monitoring and operation of the cruise control system by the remote control base station 25R via the communication device 25.

[0019] Before and during the RMF activation, the notification unit 17 notifies the occupants and passengers of the vehicle 1 and road users outside the vehicle. Furthermore, as will be described later, if the traveling direction of the vehicle 1 changes due to the activation of the RMF, the notification unit 17 may be configured to notify that the traveling direction will change. The notification unit 17 may also be configured to notify people inside and outside the vehicle that the autonomous traveling by remote monitoring and operation has stopped. The notification unit 17 may use visual methods such as turning on a lamp or displaying a text, and / or auditory methods such as outputting a voice.

[0020] The route generation unit 13 is configured to generate a target route from the vehicle position estimated by the environmental state estimation unit 11 to a destination. The route generation unit 13 searches for a route from the departure point to the destination based on the vehicle position information and map information, and generates a rough target route, a so-called global route. Furthermore, the route generation unit 13 generates a detailed target route, a so-called local route, for autonomous driving such as lane keeping, lane changes, and course changes, based on the positions of adjacent lane markings, the positions and speeds of other vehicles, and the motion state of the vehicle estimated by the environmental state estimation unit 11. The route generation unit 13 is further configured to generate a target route for RMF operation to a target stopping position when the vehicle 1 is stopped by RMF operation.

[0021] The vehicle control unit 14 calculates a target vehicle speed and a target steering angle based on the target route generated by the route generation unit 13, and transmits a speed command for constant speed driving or vehicle-to-vehicle distance maintenance / following driving to the ACC controller 15, and transmits a steering angle command for route following to the EPS controller 31 via the automatic steering controller 16.

[0022] The vehicle speed is also input to the EPS controller 31 and the ACC controller 15. Because the steering reaction force changes depending on the vehicle speed, the EPS controller 31 refers to a steering angle-steering torque map for each vehicle speed and sends a torque command to the steering mechanism 41. The engine controller 32, the ESP / ABS controller 33, and the EPS controller 31 control the engine 42, the brakes 43, and the steering mechanism 41, thereby controlling the longitudinal and lateral movements of the vehicle 1.

[0023] [Outline of the remote-controlled automated driving system] Next, an overview of the remote-controlled automated driving system will be described. The automated driving system in this embodiment is a remotely monitored and operated automated driving system, in which the traveling of vehicle 1 driven by the automated driving system is monitored by remote control base station 25R and is operated from remote control base station 25R as needed.

[0024] The autonomous driving system is a combination of an adaptive cruise control system (ACCS) and a continuous automatic steering system that automatically maintains and changes lanes continuously. The autonomous driving system can be executed when the ACC controller 15, which constitutes the adaptive cruise control system (ACC) together with the automatic driving device 10, and the automatic steering controller 16, which constitutes the continuous automatic steering system, are both operating.

[0025] An operator at the remote-controlled base station 25R sets a departure point and a destination before the vehicle 1 departs. The set departure point and destination are input from the remote-controlled base station 25R to the automatic driving device 10 of the vehicle 1 via the communication device 25. The route generation unit 13 generates a global route from the departure point to the destination based on the vehicle's position information and map information obtained from the environmental state estimation unit 11. Based on the generated global route, the route generation unit 13 generates a local route and a target vehicle speed based on external information (lanes, vehicle position, positions and speeds of other vehicles traveling in the lane in which the vehicle is traveling and adjacent lanes) obtained by the external sensor 21 and internal information (vehicle speed, yaw rate, acceleration) obtained by the internal sensor 22.

[0026] When the operator of the remote control base station 25R determines that the environmental conditions, road conditions, etc. are maintained within the system's operation design domain (ODD), he / she sends a start command to the automatic operation device 10 to cause the vehicle 1 to travel according to the generated target route and target vehicle speed.

[0027] The vehicle control unit 14 calculates the yaw rate γ and lateral acceleration (d 2 y / dt 2 ) the vehicle speed, attitude, and lateral displacement after Δt seconds are estimated from the relationship. Vehicle control unit 14 provides a steering angle command to EPS controller 31 via automatic steering controller 16 so that the lateral displacement after Δt seconds will be yt, and provides a speed command to ACC controller 15 so that the speed will be Vt after Δt seconds.

[0028] The ACC controller 15, automatic steering controller 16, EPS controller 31, engine controller 32, and ESP / ABS controller 33 operate independently of automatic steering, but can also be operated by command input from the automatic driving controller 10 while the automatic driving system is operating.

[0029] The ESP / ABS controller 33, which receives a deceleration command from the ACC controller 15, issues a hydraulic command to an actuator and controls the braking force of a brake 43, thereby controlling the vehicle speed. Furthermore, the engine controller 32, which receives an acceleration / deceleration command from the ACC controller 15, controls the actuator output (throttle opening) to issue a torque command to the engine 42, thereby controlling the driving force and thereby controlling the vehicle speed.

[0030] The ACC function (ACCS) functions as a combination of hardware and software, such as a millimeter wave radar as forward detection means 211 constituting the external sensor 21, the ACC controller 15, the engine controller 32, and the ESP / ABS controller 33.

[0031] In other words, if there is no preceding vehicle, the vehicle will travel at a constant speed with the ACC set speed (set speed) as the target vehicle speed.If the vehicle catches up with the preceding vehicle (if the preceding vehicle speed is equal to or lower than the ACC set speed), the vehicle will follow the preceding vehicle while maintaining a distance (set distance) according to the set time gap (time between vehicles = distance between vehicles / vehicle speed) in accordance with the speed of the preceding vehicle.

[0032] The continuous automatic steering system detects lane markings, the vehicle's position, and the positions and speeds of other vehicles traveling in adjacent lanes using the environmental state estimation unit 11 of the automatic operation device 10, based on image data and point cloud data acquired by external sensors 21 and vehicle information acquired by internal sensors 22. Based on this information, the continuous automatic steering system performs steering control using the EPS controller 31 via the automatic steering controller 16 to perform lane keeping control to keep the vehicle in the center of the lane and lane change control to cross lane markings.

[0033] That is, upon receiving a steering angle command from the automatic steering controller 16, the EPS controller 31 refers to a map of vehicle speed-steering angle-steering torque, issues a torque command to the actuator (EPS motor), and gives the steering mechanism 41 the target front wheel steering angle.

[0034] As described above, the autonomous driving system is a system configured by combining longitudinal control (speed control, inter-vehicle distance control) by the ACC controller 15 and lateral control (lane keeping control, lane change control) by the automatic steering controller 16. The remote-controlled autonomous driving system monitors the traveling of the vehicle 1 by the remote control base station 25R, and controls the traveling of the vehicle 1 according to remote control commands from the remote control base station 25R.

[0035] [Risk Mitigation Function (RMF) for Automated Driving Systems] While the autonomous driving system is operating, the environmental state estimation unit 11 constantly monitors whether the vehicle's running state, surrounding environmental conditions, etc. are maintained within the operation design domain (ODD) of the system, based on external information acquired through the external sensor 21, vehicle information acquired by the internal sensor 22, the communication status with the remote control base station 25R, etc. If the vehicle is outside the ODD or if the system is not operating normally, it is required to activate the RMF, which safely stops the vehicle 1 within the target stopping area.

[0036] Based on a signal from the environmental state estimation unit 11, the RMF unit 12 determines to activate the RMF when it becomes difficult for the remote control base station 25R to remotely monitor and operate the vehicle 1 due to, for example, a disconnection of communication with the remote control base station 25R or a failure of the communication device 25, and activates the RMF to guide the vehicle 1 to a target stopping area and stop it. The target stopping area for the RMF is usually set as a space on the target route for automated driving where the vehicle 1 can be stopped. However, if the RMF is activated near an intersection, there is a possibility that the vehicle 1 may cross paths with other traffic participants at the intersection or disrupt traffic flow while decelerating and guiding the vehicle 1 to the target stopping area on the target route.

[0037] The relationship between the stopping position of a vehicle and other traffic participants around an intersection will be explained with reference to Figures 4 to 6. Figures 4(a) and (b) show an example of turning right at an intersection, Figures 5(a) and (b) show an example of turning left at an intersection, and Figures 6(a) and (b) show an example of going straight through an intersection.

[0038] Figures 4(a) and (b) show an example of RMF operation near an intersection without traffic lights. Note that although traffic rules include driving on the left and driving on the right, the following explanation will use driving on the left as an example.

[0039] In the example of a crossroad shown in Figure 4(a), the target route for vehicle 1 is set to a route that turns right at intersection C1, as indicated by arrow PA. In addition to vehicle 1, there may be a vehicle 2 moving straight to the right entering intersection C1 from the right on path P2, a vehicle 3 moving straight to the left entering intersection C1 from the left on path P3, and an oncoming vehicle 4 moving straight from the oncoming lane entering intersection C1 on path P4. There are crosswalks just before intersection C1 and just beyond where vehicle 1 will turn right. In this case, there may be pedestrians attempting to cross crosswalk D1 just before intersection C1 and cross crosswalk D2 just beyond where the vehicle will turn right.

[0040] Here, when it becomes difficult to continue communication between the vehicle 1 and the remote control base station 25R near the intersection C1, the RMF unit 12 determines that it is necessary to activate the RMF.

[0041] When a target stop position S is set on the target route after turning right at intersection C1, vehicle 1 turns right along route PA due to RMF operation of automatic operation device 10, and is then guided to target stop position S along the RMF route indicated by arrow PB and stopped there. In this case, when turning right at intersection C1, vehicle 1 risks intersecting with vehicle 2 moving straight to the right in risk area RA1, with vehicle 3 moving straight to the left in risk area RA2, and with oncoming vehicle 4 moving straight in risk area RA3. In addition, vehicle 1 risks intersecting with pedestrians crossing crosswalk D1 in risk area RA4, and with pedestrians crossing crosswalk D2 in risk area RA5.

[0042] In this way, when RMF is activated near intersection C1, multiple intersections occur where there is a possibility of crossing with other traffic participants. When turning right at intersection C1, as shown in Figure 4(a), there is a risk of contact or collision with other traffic participants in multiple risk areas RA1 to RA5. In other words, a maximum of five risk areas occur based on the road structure of intersection C1.

[0043] In the example of a T-junction shown in Figure 4(b), the target route for vehicle 1 is set to a route that turns right at intersection C2, as indicated by arrow PA. In addition to vehicle 1, there may be a vehicle 2 moving straight to the right that is entering intersection C2 from the right on path P2, and a vehicle 3 moving straight to the left that is entering intersection C2 from the left on path P3, around intersection C2. Also, there are crosswalks just before intersection C2 and after the right turn, as seen from vehicle 1, and there may be pedestrians trying to cross crosswalk D1 just before intersection C2 and pedestrians trying to cross crosswalk D2 after the right turn.

[0044] Here, when it becomes difficult to continue communication between the vehicle 1 and the remote control base station 25R near the intersection C2, the RMF unit 12 determines that it is necessary to activate the RMF.

[0045] If a target stop area S is set on the target route after turning right at intersection C2, vehicle 1 turns right along route PA due to RMF activation of automatic operation device 10, and is then guided along the RMF route indicated by arrow PB to target stop position S, where it stops. In this case, when turning right at intersection C2, vehicle 1 risks intersecting with vehicle 2 moving straight to the right in risk area RA1, and with vehicle 3 moving straight to the left in risk area RA2. In addition, vehicle 1 risks intersecting with pedestrians crossing crosswalk D1 in risk area RA3, and with pedestrians crossing crosswalk D2 in risk area RA4.

[0046] In this way, when RMF is activated near intersection C2, multiple intersections occur where there is a possibility of crossing with other traffic participants. As shown in Figure 4(b), when turning right at intersection C2, there is a risk of contact or collision with other traffic participants in multiple risk areas RA1 to RA4. In other words, a maximum of four risk areas occur based on the road structure of intersection C2.

[0047] Figures 5(a) and (b) show examples of RMF operation near an intersection without traffic lights. In the case of a crossroad shown in Figure 5(a), the target route for vehicle 1 is set to turn left at intersection C1, as indicated by arrow PA. In addition to vehicle 1, there is a possibility that vehicle 2 traveling straight to the right may be present around intersection C1. As seen from vehicle 1, there are crosswalks just before intersection C1 and after the left turn, and there may be pedestrians attempting to cross crosswalk D1 just before intersection C1 and cross crosswalk D2 after the left turn.

[0048] Here, when it becomes difficult to continue communication between the vehicle 1 and the remote control base station 25R near the intersection C1, the RMF unit 12 determines that it is necessary to activate the RMF.

[0049] When a target stop position S is set on the target route after turning left at intersection C1, vehicle 1 turns left along route PA due to RMF operation of automatic operation device 10, and is then guided along the RMF route indicated by arrow PB to target stop position S and stopped there. In this case, when turning left at intersection C1, vehicle 1 risks intersecting with vehicle 2 moving straight to the right in risk area RA1. In addition, vehicle 1 risks intersecting with pedestrians crossing crosswalk D1 in risk area RA2, and with pedestrians crossing crosswalk D2 in risk area RA3.

[0050] In this way, when RMF is activated near intersection C1, multiple intersections occur where there is a possibility of crossing with other traffic participants. When turning left at intersection C1, as shown in Figure 5(a), there is a risk of contact or collision with other traffic participants in multiple risk areas RA1 to RA3. In other words, a maximum of three risk areas occur based on the road structure of intersection C1.

[0051] In the example of a T-junction shown in Figure 5(b), the target route for vehicle 1 is set to a route that turns left at intersection C2, as indicated by arrow PA. In addition to vehicle 1, there is a possibility that vehicle 2 traveling straight to the right may be present around intersection C2. Furthermore, there are crosswalks just before intersection C2 and after the left turn, as seen from vehicle 1, and there may be pedestrians attempting to cross crosswalk D1 just before intersection C2 and cross crosswalk D2 after the left turn.

[0052] Here, when it becomes difficult to continue communication between the vehicle 1 and the remote control base station 25R near the intersection C2, the RMF unit 12 determines that it is necessary to activate the RMF.

[0053] If a target stopping area S is set on the target route after turning left at intersection C2, vehicle 1 turns left along route PA due to RMF activation of automatic operation device 10, and is then guided along the RMF route indicated by arrow PB to target stopping position S, where it stops. In this case, when turning left at intersection C2, vehicle 1 risks intersecting with vehicle 2 going straight to the right in risk area RA1, with a risk of intersecting with pedestrians crossing crosswalk D1 in risk area RA2, and with pedestrians crossing crosswalk D2 in risk area RA3.

[0054] In this way, when RMF is activated near intersection C2, multiple intersections occur where there is a possibility of crossing with other traffic participants. As shown in Figure 5(b), when turning left at intersection C2, there is a risk of contact or collision with other traffic participants in multiple risk areas RA1 to RA3. In other words, a maximum of three risk areas occur based on the road structure of intersection C2.

[0055] Figures 6(a) and (b) show an example in which RMF operates near an intersection without traffic lights. In the example of a crossroad shown in Figure 6(a), the target route for vehicle 1 is set to go straight through intersection C1, as indicated by arrow PA. In addition to vehicle 1, there is a possibility that a vehicle 2 going straight to the right and a vehicle 3 going straight to the left exist around intersection C1. As seen from vehicle 1, there are crosswalks just before and beyond intersection C1, and there may be pedestrians attempting to cross crosswalk D1 just before intersection C1 and pedestrians attempting to cross crosswalk D2 just beyond intersection C1.

[0056] Here, when it becomes difficult to continue communication between the vehicle 1 and the remote control base station 25R near the intersection C1, the RMF unit 12 determines that it is necessary to activate the RMF.

[0057] When a target stop position S is set on the target route beyond intersection C1, vehicle 1 travels straight along route PA due to RMF operation of automatic operation device 10, and is then guided along the RMF route indicated by arrow PB to target stop position S, where it stops. In this case, when traveling straight through intersection C1, vehicle 1 risks intersecting with vehicle 2 traveling straight to the right in risk area RA1, and with vehicle 3 traveling straight to the left in risk area RA2. In addition, vehicle 1 risks intersecting with pedestrians crossing crosswalk D1 in risk area RA3, and with pedestrians crossing crosswalk D2 in risk area RA4.

[0058] In this way, when RMF is activated near intersection C1, multiple intersection points arise where there is a possibility of crossing with other traffic participants. When proceeding straight through intersection C1, there is a risk of contact or collision with other traffic participants in multiple risk areas RA1 to RA4, as shown in Figure 6(a). In other words, a maximum of four risk areas arise based on the road structure of intersection C1.

[0059] In the case of a T-junction shown in Figure 6(b), the target route for vehicle 1 is set to a route that goes straight through intersection C2, as indicated by arrow PA. In addition to vehicle 1, there is a possibility that right- or left-turning vehicle 5 may be present around intersection C2, entering intersection C2 on route P5 to turn right or left. Also, there are crosswalks just before and beyond intersection C2 as seen from vehicle 1, and there may be pedestrians attempting to cross crosswalk D1 just before intersection C2 and pedestrians attempting to cross crosswalk D2 just beyond intersection C2.

[0060] Here, when it becomes difficult to continue communication between the vehicle 1 and the remote control base station 25R near the intersection C2, the RMF unit 12 determines that it is necessary to activate the RMF.

[0061] If a target stop area S is set on the target route after going straight through intersection C2, vehicle 1 will go straight along route PA due to RMF operation of automatic operation device 10, and then be guided along the RMF route indicated by arrow PB to target stop position S, where it will stop. In this case, when going straight through intersection C2, vehicle 1 risks intersecting with a right-turning vehicle 5 in risk area RA1. There is also a risk of intersecting with a pedestrian crossing crosswalk D1 in risk area RA2, and a risk of intersecting with a pedestrian crossing crosswalk D2 in risk area RA3.

[0062] In this way, when RMF is activated near intersection C2, multiple intersections occur where there is a possibility of crossing with other traffic participants. As shown in Figure 6(b), when going straight through intersection C2, there is a risk of contact or collision with other traffic participants in multiple risk areas RA1 to RA3. In other words, a maximum of three risk areas occur based on the road structure of intersection C2.

[0063] In this way, when RMF is activated near an intersection, vehicle 1 will encounter intersections with vehicles going straight to the right (2), vehicles going straight to the left (3), oncoming vehicles going straight (4), and vehicles turning right or left (5) as they enter or pass through the intersection. Furthermore, particularly at intersections without traffic lights, vehicle 1 will encounter intersections with pedestrians crossing the street before or after the intersection. At such intersections, there is a risk of contact or collision between vehicle 1 and other traffic participants. Furthermore, if the RMF-based stopping position for vehicle 1 is set just before or just after the intersection, it could disrupt the traffic flow of other traffic participants.

[0064] Therefore, in this embodiment, when RMF is activated near an intersection, a target stopping position for RMF is set in an area that can reduce the risk of contact or collision between vehicle 1 and other traffic participants and further suppress the impact on the traffic flow of other traffic participants.

[0065] Here, the case where RMF operates near an intersection means, for example, a situation where, if RMF is activated to decelerate and stop vehicle 1, vehicle 1 will stop at or near the intersection, resulting in disruption of traffic flow at the intersection and the risk of contact or collision with other traffic participants. For example, if RMF is activated to decelerate vehicle 1 at a predetermined deceleration rate, but vehicle 1 is at a stage (point or time) where it is not possible to stop vehicle 1 before a no-stopping zone near an intersection, which will be described later, it is determined that RMF will operate near the intersection. In other words, if RMF is activated to decelerate vehicle 1 at a predetermined deceleration rate (for example, 4.0 m / s 2 ), a predetermined area before the intersection where the vehicle 1 will stop in a no-stop area can be defined as the vicinity of the intersection.

[0066] Note that even in the so-called "near an intersection," if the vehicle 1 is already traveling in a lane dedicated to turning right or left to follow the target route for automatic driving, or if automatic steering is occurring due to automatic driving, the vehicle may be excluded from the operation of RMF near an intersection, as described below. This is because changing the route for RMF when the vehicle is already preparing to turn right or left could increase the risk.

[0067] The setting of the target stop position for the RMF in this embodiment will be described below.

[0068] [Setting target stop position for RMF] In this embodiment, the target stop position for the RMF is set according to the following basic policy. (1) Reducing intersections with other traffic participants By setting multiple target stopping position candidates, intersections with other traffic participants are reduced. (2) Reducing the impact on traffic flow of other traffic participants By setting up a no-stopping area in a predetermined area around the intersection, including the intersection, the impact on the traffic flow of other traffic participants is reduced. (3) Determining the target stopping position by comparing risk factors The target stopping position is determined by comparing the risk factors of a plurality of target stopping position candidates.

[0069] The setting of the target stop position for the RMF will be described in detail below.

[0070] (1) Reducing intersections with other traffic participants 7(a) and (b), setting of multiple target stop position candidates that takes into consideration the reduction of intersections with other traffic participants will be described. The automated driving device 10 is configured such that the environmental state estimation unit 11 acquires the vehicle position by matching the vehicle position information from the positioning means 24 with the map information in the map information database 23, and the RMF unit 12 constantly searches for multiple target stop position candidates during automated driving, based on the vehicle position and surrounding map information, in preparation for RMF activation. The target stop position candidates for RMF may be set to positions on a route different from the global route set by the automated driving device 10, taking into consideration the reduction of intersections with other traffic participants.

[0071] Figure 7(a) shows an example of an intersection C1 that is a crossroads without traffic lights. As shown in Figure 7(a), in addition to vehicle 1 in front of intersection C1, there may be vehicle 2 going straight to the right, vehicle 3 going straight to the left, and oncoming vehicle 4 going straight around intersection C1. There may also be pedestrians attempting to cross crosswalk D1 in front of intersection C1, pedestrians attempting to cross crosswalk D2 beyond intersection C1, pedestrians attempting to cross crosswalk D3 after turning left, and pedestrians attempting to cross crosswalk D4 after turning right.

[0072] When RMF is activated near an intersection C1, the possible travel directions for the target route are going straight through the intersection C1, turning left, or turning right. Therefore, in the example shown in FIG. 7(a), the RMF unit 12 sets a target stop position candidate T1 on a route going straight through the intersection C1, and sets a target stop position candidate T2 on a route turning left through the intersection C1. The target stop position candidates T1 and T2 are set in an area such as the left side of the vehicle's lane or the shoulder, respectively. When RMF is activated, the vehicle 1 is automatically guided to and stopped at position T1 or position T2 along the route defined by the RMF indicated by arrows PB1 and PB2.

[0073] In this way, the RMF unit 12 sets the target stop position candidates T1 and T2 in areas that can be reached only by a route that goes straight or a route that turns left at the intersection C1, and does not set the target stop position candidates on routes that turn right at the intersection C1. Even if the target stop position candidates T1 and T2 are set on routes that go straight or turn left at the intersection C1, the risk area RA1 of intersection with a vehicle 2 moving straight to the right, the risk area RA2 of intersection with a vehicle 3 moving straight to the left, and the risk areas RA3 to RA5 of intersection with pedestrians remain, as shown in Figure 7(a), but by excluding the route that turns right at the intersection C1, the risk area RB1 of intersection with the vehicle 3 moving straight to the left and the oncoming vehicle 4 moving straight, and the risk area RB2 of intersection with pedestrians can be eliminated.

[0074] Figure 7(b) shows an example of a T-junction intersection C2 without traffic lights. As shown in Figure 7(b), in addition to vehicle 1 just before intersection C2, there may be vehicle 2 going straight to the right and vehicle 3 going straight to the left around intersection C2. There may also be pedestrians about to cross crosswalk D1 just before intersection C2, pedestrians about to cross crosswalk D2 after turning left, and pedestrians about to cross crosswalk D3 after turning right.

[0075] When the RMF is activated near the intersection C2, the possible travel directions for the target route PA are a left turn or a right turn at the intersection C2. Therefore, in the example shown in FIG. 7(b), the RMF unit 12 sets a target stop position candidate T1 on the route that turns left at the intersection C2. The target stop position candidate T1 is set in an area such as the left side of the vehicle's lane or the shoulder. When the RMF is activated, the vehicle 1 is automatically guided to and stopped at position T1 along the route defined by the RMF, as indicated by the arrow PB.

[0076] In this way, the RMF unit 12 sets the target stop position candidate T1 in an area that can be reached only by a route that turns left at the intersection C2, and does not set the target stop position candidate on a route that turns right at the intersection C2. Even if the target stop position candidate T1 is set on a route that turns left at the intersection C2, the risk area RA1 of an intersection with a vehicle 2 moving straight on the right and the risk areas RA2 and RA3 of an intersection with a pedestrian remain, as shown in Figure 7(b), but by excluding the route that turns right at the intersection C2, the risk area RB1 of an intersection with a vehicle 3 moving straight on the left and the risk area RB2 of an intersection with a pedestrian can be eliminated.

[0077] (2) Reducing the impact on traffic flow of other traffic participants Next, the setting of multiple target stop position candidates taking into consideration the influence of other traffic participants on the traffic flow will be described with reference to FIGS. 8(a) and 8(b).

[0078] Figure 8(a) shows an example of an intersection C1, which is a crossroads without traffic lights. As shown in Figure 8(a), a vehicle 1 is present before the intersection C1. Pedestrian crossings are provided before the intersection C1, beyond the intersection C1, and at the left and right turns. When the RMF operates near the intersection C1, the possible travel directions for the target route PA are going straight through the intersection C1, turning left, or turning right.

[0079] At intersection C1 and its vicinity, a no-stopping section (no-stopping area) A is provided, which is excluded from the search for target stopping position candidates. If vehicle 1 is stopped at intersection C1 or just before or just after it due to RMF activation, there is a possibility that the traffic flow of other traffic participants will be disrupted. Therefore, by providing no-stopping section A, the impact on the traffic flow of other traffic participants is reduced. Since no target stopping position candidates are set in no-stopping section A, the no-stopping section A can also be said to be a range excluded from the setting of the target stopping position.

[0080] The no-stopping zone A is set to an appropriate range so as to reduce the impact on the traffic flow of other traffic participants at intersection C1. For example, the no-stopping zone A at and near the intersection can be set to the area inside the edge of the intersection, the area inside the stop line at an intersection with a stop line, the area within a specified distance (e.g., 6 meters) from the edge of the intersection at an intersection without a stop line or a crosswalk, the area inside the crosswalk at an intersection without a stop line but with a crosswalk, or a no-parking zone.

[0081] Note that no prohibited direction is set ahead in the traveling direction to be excluded from the search for target stop position candidates. That is, at a crossroads intersection, it is also possible to set a target stop position for RMF in the direction of a right turn, i.e., in the traveling direction in which the vehicle 1 needs to cross an oncoming lane.

[0082] 8(a) shows an example in which an area within a predetermined distance from an intersection road edge is set as a no-stop section A. In this case, the RMF unit 12 sets a target stop position candidate T1 in a stop-possible section B beyond the no-stop section A on a route going straight through the intersection C1, sets a target stop position candidate T2 in a stop-possible section B beyond the no-stop section A on a route turning left at the intersection C1, and sets a target stop position candidate T3 in a stop-possible section B beyond the no-stop section A on a route turning right at the intersection C1. The target stop position candidates T1 to T3 are each set in an area such as the left side of the own lane or the shoulder of the road. When the RMF is activated, the vehicle 1 is automatically guided to and stopped at position T1, position T2, or position T3 along the route defined by the RMF indicated by arrows PB1, PB2, and PB3.

[0083] Fig. 8(b) shows an example in which an area within a predetermined distance from the edge of the intersection C2, a T-junction without traffic lights, is set as a no-stopping section A. As shown in Fig. 14(b), vehicle 1 is present before intersection C2, and crosswalks are provided before intersection C2 and at the left and right turns. When RMF operates near intersection C1, the possible travel directions for target route PA are the left or right turn directions at intersection C2.

[0084] In this case, the RMF unit 12 sets a target stop position candidate T1 in a stop-possible section B beyond the stop-prohibited section A on the route turning left at the intersection C2, and sets a target stop position candidate T2 in a stop-possible section B beyond the stop-prohibited section A on the route turning right at the intersection C2. The target stop position candidates T1 and T2 are set in an area such as the left side of the vehicle's lane or the shoulder. When the RMF is activated, the vehicle 1 is automatically guided to and stopped at position T1 or position T2 along the route defined by the RMF indicated by arrows PB1 and PB2.

[0085] (3) Determining the target stopping position by comparing risk factors Next, the determination of a target stop position by comparing risk factors of a plurality of target stop position candidates will be described.

[0086] As explained in (1) and (2) above, the RMF unit 12 sets multiple target stop position candidates in consideration of reducing intersections and reducing the impact on traffic flow. For each of the multiple set target stop position candidates, the RMF unit 12 calculates a risk factor RF that indicates the level of risk for the vehicle 1 from its current position to reach the target stop position candidate. The risk factor RF takes a larger value as the risk increases.

[0087] Specifically, the risk factor RF is calculated as the sum (Rf1+Rf2+Rf3) of point Rf1, which relates to the number of intersections between vehicle 1 and other traffic participants at and near the intersection, point Rf2, which relates to the predicted driving distance from vehicle 1's current position to each target stopping position candidate, and point Rf3, which relates to the predicted driving time from vehicle 1's current position to each target stopping position candidate.

[0088] The points Rf1 related to the number of intersections are calculated based on the number of intersections with other traffic participants on the route from the current position of vehicle 1 to the target stop position candidate, based on the viewpoint that the greater the number of intersections between vehicle 1 and other traffic participants, the higher the risk of contact or collision. For example, 10 points are added for one intersection, 20 points for two intersections, and 50 points for five intersections. For example, 10 points are added for each additional intersection.

[0089] The number of intersections can be determined based on, for example, map information or external information acquired by the external sensor 21. In this embodiment, when the route of the vehicle 1 crosses a driving lane, it is determined that there is a possibility of an intersection with another vehicle, regardless of whether or not there is an actual vehicle, and the number of intersections is counted. Also, when the route of the vehicle 1 crosses a crosswalk, it is determined that there is a possibility of an intersection with a pedestrian, regardless of whether or not there is a pedestrian, and the number of intersections is counted. That is, the maximum number of intersections is counted based on the vehicle's position information obtained by the positioning means 24, such as the GNSS, and the route of the vehicle 1 acquired from the map information in the map information database 23 and road conditions.

[0090] The points Rf2 related to the predicted travel distance are calculated based on the predicted travel distance from the current position of vehicle 1 to the target stop position candidate, based on the viewpoint that the longer the distance until vehicle 1 stops, the higher the risk of contact or collision. For example, 0.5 points are added for a predicted travel distance of 10 meters, 1 point for 20 meters, and 2.5 points for 50 meters, and so on. For every 10 meters of the predicted travel distance, 0.5 points are added. The predicted travel distance can be estimated, for example, based on map information of the vehicle's position and the surrounding area.

[0091] The points Rf3 related to the predicted traveling time are calculated based on the predicted traveling time from the current position of vehicle 1 to the target stopping position candidate, based on the viewpoint that the longer it takes for vehicle 1 to stop, the higher the risk of contact or collision. One point is added for every 10 seconds of predicted traveling time, such as 1 point if the predicted traveling time is 10 seconds, 2 points if it is 20 seconds, and 5 points if it is 50 seconds. For example, but not limited to, the predicted traveling time may be calculated as follows: if vehicle 1 is traveling at a speed of 40 km / h and the RMF is activated, and the deceleration is 4.0 m / s 2 After decelerating to 10 km / h, the vehicle will continue to travel at a constant speed and then gradually decelerate (for example, deceleration of 1.96 m / s) just before reaching the target stop position candidate. 2 ) and stop at the target stop position candidate.

[0092] Note that the specific values ​​of the point Rf1 related to the number of intersections, the point Rf2 related to the predicted travel distance, and the point Rf3 related to the predicted travel time are merely examples, and it is of course possible to use values ​​other than those mentioned above. However, it is preferable to weight the points so that the weights increase in the order of the point Rf1 related to the number of intersections, the point Rf3 related to the predicted travel time, and the point Rf2 related to the predicted travel distance. This takes into consideration the fact that the greater the number of intersections where there is a possibility of intersection with other traffic participants, the higher the risk, and further that the longer the travel time until stopping at the target stop position, the greater the risk is thought to be compared to when the travel distance is long.

[0093] Furthermore, if the risk factors of multiple target stop position candidates have the same value, the target stop position S is selected in the following order of priority based on the direction of travel to the target stop position candidate: a direction going straight through the intersection, a direction in which the vehicle 1 turns at the intersection without crossing the oncoming lane, and a direction in which the vehicle 1 crosses the oncoming lane and turns at the intersection.

[0094] The RMF unit 12 compares the risk factors RF calculated for a plurality of target stop position candidates, and selects as the target stop position S the target stop position candidate with the smallest risk factor RF.

[0095] With reference to FIG. 9 , the selection of a target stop position S at a T-junction C2 without traffic lights will be described. In the example shown in FIG. 9 , in addition to vehicle 1 approaching intersection C2, vehicle 2 moving straight to the right and vehicle 3 moving straight to the left may be present around intersection C2. Crosswalks are provided on the road and intersecting roads ahead of vehicle 1, and pedestrians may be present at crosswalk D1 before the intersection, crosswalk D2 after a left turn, and crosswalk D3 after a right turn. The target route PA planned by the automatic operation device 10 of vehicle 1 is a route PA that turns left or right at intersection C2. As shown in FIG. 9 , RMF unit 12 sets target stop position candidate T1 in a stop-possible section B on the route that turns left at intersection C2, sets target stop position candidate T2 in a stop-possible section B on the route that turns right at intersection C2, and sets target stop position candidate T3 in a parking space beyond target stop position candidate T2.

[0096] For target stop position candidate T1, there are three intersection points: an intersection with vehicle 2 moving straight to the right in risk area RA1, an intersection with a pedestrian attempting to cross crosswalk D1 in risk area RA2, and an intersection with a pedestrian attempting to cross crosswalk D2 in risk area RA3, so the point related to the number of intersection points is Rf1 = 30. If the predicted travel distance from the current position of vehicle 1 to target stop position candidate T1 is 100 meters and the predicted travel time is 33 seconds, then the point related to the predicted travel distance is Rf2 = 5 and the point related to the predicted travel time is Rf3 = 3.3. In this case, the risk factor RF of target stop position candidate T1 is 38.3 (= Rf1 + Rf2 + Rf3).

[0097] For target stop position candidate T2, there are four intersection points: an intersection with vehicle 2 moving straight to the right in risk area RA1, an intersection with vehicle 3 moving straight to the left in risk area RA4, an intersection with a pedestrian attempting to cross crosswalk D1 in risk area RA2, and an intersection with a pedestrian attempting to cross crosswalk D3 in risk area RA5, so the point Rf1 related to the number of intersection points is 40. If the predicted travel distance from the current position of vehicle 1 to target stop position candidate T2 is 100 meters and the predicted travel time is 33 seconds, then the point Rf2 related to the predicted travel distance is 5 and the point Rf3 related to the predicted travel time is 3.3. In this case, the risk factor RF of target stop position candidate T2 is 48.3 (= Rf1 + Rf2 + Rf3).

[0098] Similar to target stop position candidate T2, target stop position candidate T3 has four intersection points: an intersection point with vehicle 2 moving straight to the right, an intersection point with vehicle 3 moving straight to the left, and an intersection point with pedestrians crossing crosswalks D1 and D3, and the score Rf1 related to the number of intersection points is 40. If the predicted travel distance from the current position of vehicle 1 to target stop position candidate T2 is 200 meters and the predicted travel time is 69 seconds, the score Rf2 related to the predicted travel distance is 10 and the score Rf3 related to the predicted travel time is 6.9. In this case, the risk factor RF of target stop position candidate T2 is 56.9 (=Rf1+Rf2+Rf3).

[0099] The RMF unit 12 compares the risk factors RF calculated for each of the plurality of target stop position candidates T1, T2, and T3, and selects as the target stop position S the target stop position candidate T1 on the left-turn route that has the smallest risk factor RF.

[0100] Next, selection of a target stop position S at an intersection C1, which is a crossroad without traffic lights, will be described with reference to Figure 10. In the example shown in Figure 10, in addition to vehicle 1 just before intersection C1, there may be vehicle 2 moving straight to the right, vehicle 3 moving straight to the left, and oncoming vehicle 4 moving straight around intersection C1. Crosswalks are provided on the roads and intersecting roads that vehicle 1 passes through intersection C1. There may be pedestrians attempting to cross crosswalk D1 just before the intersection, pedestrians attempting to cross crosswalk D2 beyond the intersection, pedestrians attempting to cross crosswalk D3 where a left turn is to be made, and pedestrians attempting to cross crosswalk D4 where a right turn is to be made. The target route PA planned by automatic operation device 10 of vehicle 1 is a route PA that goes straight, turns left, or turns right through intersection C1.

[0101] 10, the RMF unit 12 sets a target stop position candidate T1 in a stop possible section B on a route going straight through the intersection C1, and sets a target stop position candidate T2 in a stop possible section B further ahead of the target stop position candidate T1. Also, it sets a target stop position candidate T3 in a stop possible section B on a route turning left at the intersection C1, and sets a target stop position candidate T4 in a stop possible section B on a route turning right at the intersection C1.

[0102] For target stop position candidate T1, there are four intersection points: an intersection point with vehicle 2 moving straight to the right in risk area RA1, an intersection point with vehicle 3 moving straight to the left in risk area RA2, an intersection point with a pedestrian attempting to cross crosswalk D1 in risk area RA3, and an intersection point with a pedestrian attempting to cross crosswalk D2 in risk area RA4. Therefore, the score Rf1 related to the number of intersection points is 40. If the predicted travel distance from the current position of vehicle 1 to target stop position candidate T1 is 100 meters and the predicted travel time is 33 seconds, the score Rf2 related to the predicted travel distance is 5 and the score Rf3 related to the predicted travel time is 3.3. In this case, the risk factor RF of target stop position candidate T1 is 48.3 (= Rf1 + Rf2 + Rf3).

[0103] As with target stop position candidate T1, target stop position candidate T2 has four intersection points: an intersection point with vehicle 2 moving straight to the right, an intersection point with vehicle 3 moving straight to the left, and an intersection point with pedestrians crossing crosswalks D1 and D2. Therefore, the score Rf1 related to the number of intersection points is 40. If the predicted travel distance from the current position of vehicle 1 to target stop position candidate T2 is 200 meters and the predicted travel time is 69 seconds, the score Rf2 related to the predicted travel distance is 10 and the score Rf3 related to the predicted travel time is 6.9. In this case, the risk factor RF of target stop position candidate T2 is 56.9 (= Rf1 + Rf2 + Rf3).

[0104] For target stop position candidate T3, there are three intersection points: an intersection with vehicle 2 moving straight to the right in risk area RA1, an intersection with a pedestrian attempting to cross crosswalk D1 in risk area RA3, and an intersection with a pedestrian attempting to cross crosswalk D3 in risk area RA5, so the point Rf1 related to the number of intersection points is 30. If the predicted travel distance from the current position of vehicle 1 to target stop position candidate T3 is 100 meters and the predicted travel time is 33 seconds, then the point Rf2 related to the predicted travel distance is 5 and the point Rf3 related to the predicted travel time is 3.3. In this case, the risk factor RF of target stop position candidate T1 is 38.3 (= Rf1 + Rf2 + Rf3).

[0105] For target stop position candidate T4, there are five intersection points: an intersection with vehicle 2 moving straight to the right in risk area RA1, an intersection with vehicle 3 moving straight to the left and an intersection with oncoming vehicle 4 in risk area RA6, an intersection with a pedestrian attempting to cross crosswalk D1 in risk area RA3, and an intersection with a pedestrian attempting to cross crosswalk D4 in risk area RA7, so the score Rf1 related to the number of intersection points is 50. If the predicted travel distance from the current position of vehicle 1 to target stop position candidate T4 is 100 meters and the predicted travel time is 33 seconds, then the score Rf2 related to the predicted travel distance is 5 and the score Rf3 related to the predicted travel time is 3.3. In this case, the risk factor RF of target stop position candidate T1 is 58.3 (= Rf1 + Rf2 + Rf3).

[0106] The RMF unit 12 compares the risk factors RF calculated for each of the plurality of target stop position candidates T1 to T4, and selects as the target stop position S the target stop position candidate T3 on the left turn route that has the smallest risk factor RF.

[0107] [RMF operation flow during autonomous driving] Next, the flow of RMF operation when a situation arises in which remote monitoring and operation is difficult during autonomous driving will be described. Fig. 11 shows a flowchart illustrating the flow of RMF operation according to this embodiment.

[0108] (1) Automatic driving operation by remote monitoring and operation (Step S100) In response to a start command from the operator of the remote control base station 25R, automatic traveling is performed by the automatic traveling device 10. The automatic traveling device 10 controls the traveling of the vehicle 1 in accordance with the generated target route and target vehicle speed.

[0109] (2) Setting multiple target stop position candidates (step S101) The RMF unit 12 searches for target stop position candidates based on the vehicle's own position information from the positioning means 24 and the map information in the map information database 23, and sets multiple target stop position candidates along the global route to the destination in preparation for RMF activation. The RMF unit 12 sets multiple target stop position candidates at all times (for example, at predetermined intervals) during automated driving, not just near intersections. The multiple target stop position candidates are constantly updated, and the latest target stop position candidates based on the current position of the vehicle 1 are set.

[0110] (3) Communication failure determination (step S110) During operation of the automatic operation device 10 by remote monitoring and operation, the RMF unit 12 constantly (for example, at predetermined intervals) determines whether communication with the remote control base station 25R via the communication device 25 is normal and whether automatic driving can be continued. If it is determined that a communication failure such as a failure of the communication device 25 or a cutoff in communication with the remote control base station 25R has made it difficult to continue remote monitoring and operation of automatic driving by the automatic operation device 10, the process proceeds to step S111 to activate the RMF. On the other hand, if no communication failure has occurred and remote monitoring and operation can be continued, the process returns to step S101 and automatic driving continues.

[0111] (4) Remote monitoring and operation stop (steps S111 to S113) If it is determined that it has become difficult to continue remote monitoring and operation, a flag indicating a communication failure is set (step S111), and at the same time, the alarm unit 17 notifies those inside and outside the vehicle that the operation (automatic driving) of the automatic operation device 10 through remote monitoring and operation will be stopped (step S112), and the operation of the automatic operation device 10 through remote monitoring and operation will be stopped.

[0112] (5) RMF operation notification (step S120) When the RMF is activated due to a situation where it is difficult to continue remote monitoring and operation, the notification unit 17 notifies the crew and passengers of the vehicle 1. The notification unit 17 may also be configured to notify the outside of the vehicle.

[0113] (6) Selecting a target stop position (step S121) A target stop position S is selected from a plurality of target stop position candidates. The RMF unit 12 calculates the risk factor RF for each of the plurality of target stop position candidates set in step S101 as described above, and selects the target stop position candidate with the lowest risk factor RF as the target stop position S. The route generation unit 13 generates a target route for RMF operation to the selected target stop position S based on the host vehicle position information, map information, adjacent lane marking positions, other vehicle positions and speeds, the vehicle motion state, etc. acquired by the environmental state estimation unit 11.

[0114] (7) RMF operation (step S122) The RMF starts operating to safely stop vehicle 1 within the target stopping area. The RMF is a function that drives vehicle 1 under its own power to the target stopping position S according to the target route for RMF operation, and then decelerates and stops it.

[0115] (8) Determining whether or not the vehicle can stop at the target stop position S (step S128) The RMF unit 12 detects the presence or absence of obstacles around the target stopping position S based on obstacle information around the target stopping position S provided from the external sensor 21 via the environmental state estimation unit 11, and determines whether or not it is possible to stop at the target stopping position S. If no obstacles are present around the target stopping position S, it determines that stopping is possible and proceeds to step S130.

[0116] On the other hand, for example, if there is another vehicle at the target stopping position S or if construction work is being carried out near the target stopping position S, it is determined that the vehicle 1 cannot stop at the target stopping position S due to obstacles around the target stopping position S. In this case, the process returns to step S121 and a new target stopping position S is selected.

[0117] (9) Deceleration and Stopping by RMF (Steps S130 to S132) The automatic operation device 10 activates a direction indicator (not shown) in the direction of the target stop position S (step S130). The vehicle control unit 14 performs speed control and steering control so that the vehicle 1 travels while decelerating according to the target route for RMF operation, and stops the vehicle at the target stop position S (step S131). The automatic operation device 10 flashes hazard lights (not shown) (step S132).

[0118] (10) RMF Completion Determination (Steps S133 to S134) The RMF unit 12 matches the position where the vehicle 1 has stopped with the target stop position S based on the vehicle position information and map information, and determines whether they match. If they do not match, the process returns to step S131 and the vehicle travels to the target stop position S. If the stop position of the vehicle 1 matches the target stop position S, the process determines that the RMF has been completed, and proceeds to step S134, where the RMF is stopped. This ends the series of RMF processes.

[0119] The driving control device for the vehicle 1 according to the present embodiment described above can achieve the following advantageous effects.

[0120] A cruise control device of a vehicle 1 equipped with an automatic operation device 10 for performing automatic driving through remote monitoring and operation has a risk mitigation function (RMF) that performs risk mitigation control to stop the vehicle 1 at a target stop position when it becomes difficult to continue remote monitoring and operation. The automatic operation device 10 is configured to search for target stop position candidates based on position information and map information of the vehicle 1 in preparation for activation of the RMF during automatic driving, and when the RMF is activated near an intersection, calculate a risk factor RF that represents the level of risk involved in the vehicle 1 reaching the target stop position candidate from its current position for each of multiple target stop position candidates, including target stop position candidates located on routes different from the target route PA of the automatic driving, and select the target stop position candidate with the lowest risk factor RF from the multiple target stop position candidates as the target stop position S.

[0121] Since multiple target stop position candidates are searched for during automatic driving, the target stop position S can be selected quickly when RMF operation begins. When RMF operates near an intersection, the target stop position with the lowest risk factor can be selected by comparing the risk factors RF of each target stop position candidate, minimizing the risk of disrupting traffic flow and inducing contact or collision with other traffic participants.

[0122] The risk factor RF is calculated as the sum of point Rf1, which relates to the number of intersections between vehicle 1 and other traffic participants at and near the intersection; point Rf3, which relates to the predicted travel time from vehicle 1's current position to the candidate target stop position; and point Rf2, which relates to the predicted travel distance from vehicle 1's current position to the candidate target stop position. Regarding the number of intersections, the fewer the number of intersections with other traffic participants, the lower the possibility of contact or collision. Regarding the predicted travel time, the shorter the travel time to the target stop position, the shorter the travel time when it is difficult to continue remote monitoring and operation. Regarding the predicted travel distance, the shorter the travel distance to the target stop position, the shorter the travel time when it is difficult to continue remote monitoring and operation. Therefore, an appropriate target stop position can be determined from the perspective of reducing intersections with other traffic participants when passing through an intersection and reducing the travel time and distance to the target stop position. In this embodiment, regardless of whether other traffic participants are actually present or not, the number of intersections is counted from the road structure based on the vehicle position information obtained from the positioning means 24 such as GNSS and the map information in the map information database 23, so it is also possible to reduce the processing load in the RMF unit 12.

[0123] The risk factor RF is calculated by weighting the points Rf1 related to the number of intersections, Rf3 related to the predicted travel time, and Rf2 related to the predicted travel distance in that order. For example, because it is expected that external influences such as roadside construction will cause the required time to reach the target stopping position even if the distance is short, the predicted travel time is weighted more heavily than the predicted travel distance.

[0124] When multiple target stop position candidates have the same risk factor RF, the direction of travel to the target stop position candidate is selected as the target stop position S in the following order of priority: a direction of going straight through the intersection, a direction in which the vehicle 1 turns at the intersection without crossing the oncoming lane, and a direction in which the vehicle 1 crosses the oncoming lane and turns at the intersection. When the target stop position S is set in the straight direction, the external sensor 21 can quickly detect the presence or absence of obstacles around the target stop position S and determine whether or not it is possible to stop there, so the target stop position candidate in the straight direction is selected preferentially. With regard to the direction in which the vehicle 1 turns at an intersection without crossing the oncoming lane (for example, a left turn), and the direction in which the vehicle 1 crosses the oncoming lane and turns at the intersection (for example, a right turn), it is considered that a left turn has no intersection with an oncoming vehicle 4 going straight, and therefore has a lower risk than a right turn.

[0125] At intersections and near intersections, no-stopping areas A are set that are excluded from the search for target stop position candidates. High-risk areas near intersections are set as no-stopping areas A, and a target stop position S is not set in the no-stopping areas A. This makes it possible to at least partially eliminate intersections with other traffic participants and reduce the risk of contact or collision with other traffic participants.

[0126] -Variations- (1) In the above-described embodiment, in preparation for RMF activation, a plurality of target stop position candidates are constantly searched for, and the target stop position S is selected from the plurality of target stop position candidates. Since it is anticipated that an obstacle may exist around the target stop position S, making it impossible to stop, it is preferable to have a large number of target stop position candidates. From the viewpoint of the processing load on the RMF unit 12, the number of target stop position candidates may be set to at least two, and if it is impossible to stop at the selected target stop position S, a target stop position candidate located on the same route may be added.

[0127] (2) In the above-described embodiment, the determination of the target stop position S when the RMF is activated near an intersection has been described. However, the method for determining the target stop position S according to the above-described embodiment can be applied not only to the vicinity of an intersection but also to situations where the activation of the RMF requires a lane change or a course change, such as on a multi-lane road.

[0128] (3) In the above-described embodiment, the risk factor RF is calculated as the sum of the point Rf1 related to the number of intersections, the point Rf2 related to the predicted travel distance, and the point Rf3 related to the predicted travel time. However, this is not limited to this, and the risk factor RF may be calculated using at least one of the point Rf1 related to the number of intersections, the point Rf2 related to the predicted travel distance, and the point Rf3 related to the predicted travel time. Furthermore, the risk factor RF may be calculated based on factors other than the number of intersections, the predicted travel distance, and the predicted travel time.

[0129] (4) In the above-described embodiment, the traffic rule is to keep to the left, but this embodiment can also be applied to cases where the traffic rule is to keep to the right. In cases where the traffic rule is to keep to the right, turning right at an intersection means turning in a direction that does not cross the oncoming lane, and turning left at an intersection means turning in a direction that crosses the oncoming lane.

[0130] Although several embodiments of the present invention have been described above, it should be noted that the present invention is not limited to the above-described embodiments, and various modifications and variations are possible within the scope of the present invention. [Explanation of symbols]

[0131] 1 vehicle 10 Automatic operation device 11 Environmental state estimation unit 12 RMF section 13 Route generation unit 14 Vehicle control unit 15 ACC controller 16 Autopilot Controller 17. Information Department 21 External Sensor 22 Internal Sensors 23 Map Information Database 24 Positioning Method (GNSS) 25 Communication Device 31 EPS controller 32 Engine Controller 33 ESP / ABS controller 34 Manual steering (steering wheel) 35 Manual operation (accelerator pedal) 36 Manual operation (brake pedal)

Claims

1. A vehicle driving control device equipped with an automatic driving device for performing automatic driving by remote monitoring and operation, A risk mitigation function (RMF) that executes risk mitigation control to stop the vehicle at a target stopping position when it becomes difficult to continue remote monitoring and operation, The automatic operation device is During the automatic driving, a search for a plurality of target stopping position candidates is always performed based on the position information and map information of the vehicle in preparation for the activation of the RMF; When the RMF is activated near an intersection, a risk factor is calculated for each of a plurality of target stop position candidates, including target stop position candidates located on a route different from the target route of the automated driving, to indicate a level of risk for the vehicle to reach the target stop position candidate from its current position; a target stop position candidate having the lowest risk factor is selected as the target stop position from among the plurality of target stop position candidates, A vehicle driving control device, wherein the risk factor is calculated as the sum of points related to the number of intersections between the vehicle and other traffic participants at the intersection and near the intersection, points related to the predicted travel time from the vehicle's current position to a candidate target stopping position, and points related to the predicted travel distance from the vehicle's current position to a candidate target stopping position.

2. 2. The vehicle cruise control device according to claim 1, wherein the risk factor is calculated by weighting the points related to the number of intersections, the points related to the predicted travel time, and the points related to the predicted travel distance in that order.

3. 3. A vehicle driving control device according to claim 1, wherein, when the risk factors of the plurality of target stop position candidates are the same, the target stop position is selected in the following order of priority as a direction of travel to the target stop position candidate: a direction of going straight through the intersection, a direction in which the vehicle turns at the intersection without crossing an oncoming lane, and a direction in which the vehicle crosses the oncoming lane and turns at the intersection.

4. The vehicle cruise control device according to claim 1 , wherein a no-stop area is set at the intersection and in the vicinity of the intersection, the no-stop area being excluded from a search for target stop position candidates.

Images